A method for analyzing the probability failure risk of life-limited parts considering random load interference
By constructing a stochastic load history and crack propagation model throughout the entire life cycle, the inaccuracy of failure risk assessment for life-limited components in existing technologies is solved, and efficient and accurate failure risk analysis in complex environments is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIHANG UNIV
- Filing Date
- 2022-09-02
- Publication Date
- 2026-07-24
AI Technical Summary
Existing probabilistic failure analysis methods are not suitable for analyzing complex environments, and the analysis results are inaccurate, especially under random load interference of life-limited components of aero-engines, making it difficult to achieve accurate failure risk assessment.
By conducting finite element analysis on life-limited components of aero-engines, a random load history throughout the entire life cycle is constructed. Combining load randomness indices and crack propagation models, an interval characterization probability method is used to calculate crack propagation and assess failure risk.
It enables accurate failure probability analysis of life-limited components in complex environments, improving the efficiency and reliability of the analysis and supporting full life-cycle analysis under digital twins.
Smart Images
Figure CN115455766B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of structural component risk assessment technology, and relates to a method for probabilistic failure risk analysis of life-limited components that considers random load interference. Background Technology
[0002] The safety of aero-engines is a critical issue in flight safety. Since the primary failure of life-limited components in aero-engines can have catastrophic consequences, ensuring the safety of these components is a key aspect of engine safety. To address the safety of life-limited components, the aviation industry has developed a life-limited component assessment process centered on failure probability risk analysis.
[0003] Among these methods, probabilistic failure analysis is a crucial step in the probabilistic failure risk assessment of life-limited components. Traditional failure analysis methods, such as the equivalent load method, first- / second-order reliability methods, and numerical integration methods, all treat the load spectrum under different cycles throughout the entire life cycle as a specific load, and then perform crack propagation calculations. When the cycles change from the standard state, the interference effects between cycles cannot be analyzed using these methods. Furthermore, when the real-time flight conditions change, the corresponding risks cannot be updated in real time using these methods. If the Monte Carlo method is used to analyze the above problems, it will result in low analysis efficiency due to the need for sampling on the order of millions.
[0004] In summary, current traditional failure probability analysis methods are not suitable for analyzing complex environments, and the analysis results are inaccurate. Summary of the Invention
[0005] In view of the above analysis, the present invention provides a method for probabilistic failure risk analysis of life-limited components that considers random load interference, in order to solve the problems that existing probabilistic failure analysis methods cannot adapt to complex environments and that the analysis results are inaccurate.
[0006] This invention provides a method for probabilistic failure risk analysis of life-limited components considering random load interference, comprising the following steps:
[0007] Step 1: Perform finite element analysis on the working environment of the life-limited component of the aero-engine under typical flight cycles to obtain the single-cycle load value at the finite element analysis node of the life-limited component.
[0008] Step 2: Determine the single-cycle characteristic load based on the single-cycle load value; select the load randomness index; construct the load history of all cycles throughout the entire life cycle of the life-limited component using the single-cycle characteristic load and the load randomness index;
[0009] Step 3: Obtain the transcendence curve of the initial crack distribution of the component material with limited lifespan, transform the transcendence curve into a probability distribution in the form of intervals, and mark the sequence number of each interval, the initial crack length at the endpoint of each interval, and the probability that the initial crack length at the endpoint falls into the corresponding interval.
[0010] Step 4: Select a crack propagation model based on load interference effect, and based on the load history of all cycles in the whole life cycle in Step 2, calculate the crack propagation of the initial crack length at the endpoint of each interval to obtain the final value of crack propagation at the endpoint of each interval.
[0011] Step 5: Match the final crack propagation value at each interval endpoint with the probability that the initial crack length at the endpoint falls within the corresponding interval to obtain the final crack propagation value at each interval endpoint and the probability distribution of the final crack propagation value at each interval endpoint falling within the corresponding interval.
[0012] Step 6: Compare the selected judgment criteria with the final crack propagation value at each interval endpoint and the probability distribution of the final crack propagation value at each interval endpoint falling into the corresponding interval to determine the probabilistic failure risk after the entire life cycle.
[0013] Optionally, step one specifically includes: obtaining the temperature, flow rate, pressure, and rotational speed of a typical flight cycle lower-life component; substituting the temperature, flow rate, pressure, and rotational speed into ANSYS CFX for fluid analysis to obtain the volume temperature distribution data of the lower-life component; and substituting the volume temperature distribution data and rotational speed data into the Transient Structural module of ANSYS WORKBENCH to obtain the single-cycle load values on all finite element nodes of the single-cycle lower-life component.
[0014] Optionally, step two specifically includes: selecting the maximum value s of the single-cycle load at the finite element node. base As a single-cycle characteristic load; the load randomness index includes dispersion, random load distribution pattern and number of cycles in the whole life cycle determined according to working history data or airworthiness standards; the load history of all cycles in the whole life cycle is constructed by random sampling.
[0015] Optionally, the random load s is obtained based on the single-cycle characteristic load and dispersion; the load distribution law within the entire cycle is obtained based on the single-cycle characteristic load and random distribution law.
[0016] Optionally, step three specifically includes: selecting the transcendence curve of the initial crack distribution of the life-limited component; dividing the size range of the initial crack distribution transcendence curve into x intervals at equal intervals; labeling each x interval with an interval number d, where d = 1, 2, 3, ..., x; and recording the crack length at the left endpoint of each interval after equal interval division. and the length of the crack at the right end Calculate the crack length at the left end of each interval. and the length of the crack at the right end The probability of falling in interval d
[0017] Optionally, step four specifically includes: selecting the Wheeler load interference model with the built-in Paris formula, and obtaining the crack length at the right end of each interval based on the load history of the entire life cycle obtained in step two. and the length of the crack at the right end Correspondingly, the final value of crack length propagation at the left end of each interval after the full life cycle loading history. and the final value of the crack length propagation at the right end
[0018] Optionally, step five specifically includes: calculating the probability that the final value of the crack length extension at each interval endpoint falls within interval d based on the principle of probability conservation. middle.
[0019] Optionally, step six specifically includes: selecting the fracture toughness K. c As a criterion, the crack length a is obtained. critical ;
[0020] Extend the final value of the crack length at the left and right endpoints of each interval. Determining the crack length a critical Compared to, obtaining satisfaction The interval number corresponding to the condition is used to generate the failure risk interval number j; if Then the failure risk P f =0, there is no failure risk interval index j;
[0021] If an interval index j exists, then the total failure risk P f for:
[0022]
[0023] in, Let x be the probability value corresponding to the k-th interval, and let x be the maximum value of the interval index.
[0024] Optionally, the life-limiting component is the compressor wheel.
[0025] Optionally, the compressor disc is made of titanium alloy.
[0026] Compared with the prior art, the present invention can achieve the following beneficial effects:
[0027] 1) Based on the load history and load randomness of the entire life cycle, this invention constructs a random complete load history of the entire life cycle, which makes the analysis of the failure probability of life-limited components more accurate.
[0028] 2) This invention analyzes the failure probability of life-limited components based on random load interference and crack propagation, which can reflect the actual crack propagation situation, thereby enhancing the reliability of the failure risk analysis method at the load analysis level.
[0029] 3) Based on the probability density evolution theory of cracks, this invention can realize that the risk is updated with the update of the working load, and can support the full life cycle analysis of life-limited components under digital twin.
[0030] 4) The interval representation probability-based method used in this invention, compared with the Monte Carlo method, can avoid a large number of samples in the analysis, thus ensuring higher analysis efficiency under complex load environments. Attached Figure Description
[0031] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.
[0032] Figure 1 This is a flowchart of the failure probability risk analysis method for life-limited components of the present invention;
[0033] Figure 2 This is the transcendence curve of the initial crack distribution in this invention;
[0034] Figure 3 This is a probability distribution diagram of the initial crack distribution between partitions in this invention. Detailed Implementation
[0035] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.
[0036] A specific embodiment of the present invention, such as Figure 1-3 A method for probabilistic failure risk analysis of life-limited components considering random load interference is disclosed, including the following steps:
[0037] Step 1: Perform finite element analysis on the working environment of the life-limited component of the aero-engine under typical flight cycles to obtain the single-cycle load value at the finite element analysis node of the life-limited component.
[0038] Step 2: Determine the single-cycle characteristic load based on the single-cycle load value; select the load randomness index; construct the load history of all cycles throughout the entire life cycle of the life-limited component using the single-cycle characteristic load and the load randomness index;
[0039] Step 3: Obtain the transcendence curve of the initial crack distribution of the component material with limited lifespan, transform the transcendence curve into a probability distribution in the form of intervals, and mark the sequence number of each interval, the initial crack length at the endpoint of each interval, and the probability that the initial crack length at the endpoint falls into the corresponding interval.
[0040] Step 4: Select a crack propagation model based on load interference effect, and based on the load history of all cycles in the whole life cycle in Step 2, calculate the crack propagation of the initial crack length at the endpoint of each interval to obtain the final value of crack propagation at the endpoint of each interval.
[0041] Step 5: Match the final crack propagation value at each interval endpoint with the probability that the initial crack length at the endpoint falls within the corresponding interval to obtain the final crack propagation value at each interval endpoint and the probability distribution of the final crack propagation value at each interval endpoint falling within the corresponding interval.
[0042] Step 6: Compare the selected judgment criteria with the final crack propagation value at each interval endpoint and the probability distribution of the final crack propagation value at each interval endpoint falling within the corresponding interval to determine the failure probability risk after the entire life cycle.
[0043] Optionally, step one specifically includes: acquiring the temperature, flow rate, pressure, and rotational speed of a typical flight cycle lower-life component. Substituting the temperature, flow rate, pressure, and rotational speed into ANSYS CFX for fluid analysis to obtain the volumetric temperature distribution data of the life-limiting component. Substituting the volumetric temperature distribution data and rotational speed data into the Transient Structural module of ANSYS WORKBENCH to obtain the single-cycle load values at all finite element nodes of the single-cycle lower-life component. Preferably, the life-limiting component is a compressor disk.
[0044] Optionally, step two specifically includes: selecting the maximum value s of the single-cycle load at the finite element node. base The load is a single-cycle characteristic load; the load randomness index includes dispersion, random load distribution law and the number of cycles in the whole life cycle determined by working history data or airworthiness standards; the random load s is obtained based on the single-cycle characteristic load and dispersion; the load distribution law in the whole cycle is obtained based on the single-cycle characteristic load and random distribution law; the load history of all cycles in the whole life cycle is constructed by random sampling.
[0045] Preferably, the maximum value s of the single-cycle load value of all finite element nodes of the life-limited component in step one is selected. base As a single-cycle characteristic load; the dispersion is determined according to airworthiness standards; the random load s is obtained based on the single-cycle characteristic load and the dispersion; the normal distribution is selected as the distribution law of the random load, based on the 3σ principle of the normal distribution (s base -3×σ, s base +3×σ) is used to characterize the random load s; based on the single-cycle characteristic load and the random distribution law, the load distribution law s~U(s) within the entire cycle period is obtained. base , σ 2Based on historical work data, the number of cycles i for the entire life cycle is selected, i.e., i takeoff and landing cycles are performed within the entire life cycle of the life-limited component; the load distribution pattern s~U(s) within the entire cycle is analyzed. base , σ 2 The load value s is obtained by random sampling. i , where s i Let be the load value in the i-th cycle, and let be the load value s. i The samples are combined in the order of sampling to obtain the load history of the limited-life component for all cycles throughout its entire life cycle, wherein the total number of samplings is equal to the number of cycles i in the entire life cycle.
[0046] In this example, s base =620 MPa, dispersion is 20%, at this time the random load s = (s base -20%×s base s base +20%×s base )=(496,744); The variance σ=41.3 in the normal distribution of the random load distribution law is (496,743.9), and (496,743.9) is used to characterize the interval (496,744); The load distribution law within the whole cycle period is obtained from the single-cycle characteristic load and the random distribution law as s~U(620,41.3) 2 Based on historical work data, the cycle number i = 20000 for the entire lifespan was selected. Random sampling was performed on the above normal distribution using Matlab software, with the total number of samplings equal to the cycle number i for the entire lifespan. Load values s were obtained after sampling the load distribution within the above full lifespan. i ,; by using the load value s i By combining the samples in the order of sampling, the load histories of the compressor disk for all cycles throughout its entire lifespan are obtained as s1, s2...s 20000 .
[0047] Optionally, step three specifically includes:
[0048] Select the initial crack distribution exceedance curve of the component material in FAA Airworthiness Advisory Circular AC33.14. Divide the dimensional range containing the exceedance curve of the initial crack distribution into x intervals at equal intervals. Label each x interval with an interval number d, where d = 1, 2, 3, ..., x. Record the crack length at the left endpoint of each interval after equal interval division. and the length of the crack at the right end Calculate the crack length at the left end of each interval. and the length of the crack at the right end The probability of falling in interval d
[0049]
[0050] Where N() is the transcendence number in the transcendence curve of the initial crack distribution, and α min α max These represent the minimum and maximum crack lengths in the transcendental curve of the initial crack distribution.
[0051] Preferably, the transcendence curve of the initial defects of TC4 titanium alloy in FAA Airworthiness Advisory Circular AC33.14 is selected as the transcendence curve of the initial crack distribution of the compressor disk component material (TC4 titanium alloy material) in this example. See [link to relevant documentation]. Figure 2 The exceedance curve includes the finite element nodes corresponding to the maximum load, representing the situation where cracks appear at any location on the compressor disk.
[0052] Through preliminary experiments, the transcendence curve of the initial crack distribution was divided into 50 equally spaced intervals, a min =1.46×10 -4 m, a max =2.07×10 -4 m, thus obtaining the initial value distribution information of crack propagation, Table 1.
[0053] Table 1 Initial value distribution information for crack propagation
[0054]
[0055] Optionally, step four specifically includes: based on the load history of the entire life cycle obtained in step two, under the action of load, the crack generates plastic zones of different sizes at the crack tip, which in turn affects the subsequent propagation characteristics, that is, the interference effect between loads is generated.
[0056] First, the Wheeler load interference model with built-in Paris formula is selected to calculate the crack length 'a' after each cycle. i+1 :
[0057]
[0058] in, C represents the crack propagation rate in each cycle. p ΔK is the correction coefficient for the plastic zone at the crack tip, used to characterize the interference effect in the plastic zone, and can be obtained from equation (3). C and n are the coefficients of the Paris formula, obtained by experimental fitting under constant amplitude load tension, and π is pi. i a i and s i Let a be the stress intensity factor, crack length, and total load value for the entire life cycle at the i-th cycle. i+1Let G be the crack length in the (i+1)th cycle; G is the shape factor, which is 0.6371 according to the Newman method.
[0059]
[0060] Among them, a p The sum of crack length and plastic zone size at the moment when the historically largest plastic zone is generated; m is the Wheeler formula coefficient, the larger m is, the more significant the interference effect and the lower the crack propagation rate; ΔK i and a i These represent the stress intensity factor and crack length during the i-th full life cycle, respectively; r i Let be the length of the plastic zone during the i-th full-life cycle. It is calculated using the following formula:
[0061]
[0062] Where π is the mathematical constant pi, and σ ys Let ΔK be the yield strength of the material. i The stress intensity factor is the stress intensity factor when the i-th full life cycle is completed.
[0063] By using the above formula, the crack length at the right end of each interval is obtained by traversing the load history throughout the entire life cycle. and the length of the crack at the right end Correspondingly, the final value of crack length propagation at the left end of each interval after the full life cycle loading history. and the final value of the crack length propagation at the right end
[0064] Preferably, the material of the wheel is a titanium alloy, and for titanium alloy, 1.3≤m≤2, preferably, m=1.3.
[0065] Optionally, step five specifically includes: obtaining the probability that the final value of the crack length propagation at the endpoint of each interval falls within interval d, based on the principle of probability conservation.
[0066]
[0067] Where f(a) is the probability density function of the initial crack distribution. Given the initial value of crack propagation, the probability that the crack length at the left end and the crack length at the right end of each interval fall within interval d; Given the final crack propagation value, this represents the probability that the final crack length propagation values at the left and right endpoints of each interval fall within interval d. Based on this, Table 2 provides the distribution information for the final crack propagation values.
[0068] Table 2. Crack propagation final value distribution information
[0069]
[0070] Optionally, step six specifically includes: selecting the fracture toughness K. c As a criterion, the crack length a is obtained. critical :
[0071]
[0072] Among them, K c π is the fracture toughness, G is the shape factor, and s is the circumference ratio. base The maximum value of a single-cycle load on a finite element node of a finite element with a limited lifespan.
[0073] Expand the final values of the crack length at the left and right endpoints of each interval in Table 2. Determining the crack length a critical Compared to, obtaining satisfaction The interval number corresponding to the condition is used to obtain the failure risk interval number j. Then the failure risk P f =0, there is no failure risk interval index j.
[0074] If an interval index j exists, then the total failure risk P f for:
[0075]
[0076] in, Let x be the probability value corresponding to the k-th interval, and let x be the maximum value of the interval index. In this example, x = 50.
[0077] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for probabilistic failure risk analysis of life-limited components considering random load interference, characterized in that, Includes the following steps: Step 1: Perform finite element analysis on the working environment of the life-limited component of the aero-engine under typical flight cycles to obtain the single-cycle load value at the finite element analysis node of the life-limited component. Step 2: Determine the single-cycle characteristic load based on the single-cycle load value; Select load randomness index; construct the load history of all cycles throughout the life cycle of the life-limited component using single-cycle characteristic load and load randomness index; Step 3: Obtain the transcendence curve of the initial crack distribution of the component material with limited lifespan, transform the transcendence curve into a probability distribution in the form of intervals, and mark the sequence number of each interval, the initial crack length at the endpoint of each interval, and the probability that the initial crack length at the endpoint falls into the corresponding interval. Step 4: Select a crack propagation model based on load interference effect, and based on the load history of all cycles in the whole life cycle in Step 2, calculate the crack propagation of the initial crack length at the endpoint of each interval to obtain the final value of crack propagation at the endpoint of each interval. Step 5: Match the final crack propagation value at each interval endpoint with the probability that the initial crack length at the endpoint falls within the corresponding interval to obtain the final crack propagation value at each interval endpoint and the probability distribution of the final crack propagation value at each interval endpoint falling within the corresponding interval. Step 6: Compare the selected judgment criteria with the final crack propagation value at each interval endpoint and the probability distribution of the final crack propagation value at each interval endpoint falling into the corresponding interval to determine the probabilistic failure risk after the entire life cycle.
2. The method for probabilistic failure risk analysis of life-limited components according to claim 1, characterized in that, Step one specifically includes: obtaining the temperature, flow rate, pressure, and rotational speed of a typical flight cycle lower-life component; substituting the temperature, flow rate, pressure, and rotational speed into ANSYS CFX for fluid analysis to obtain the volume temperature distribution data of the lower-life component; and substituting the volume temperature distribution data and rotational speed data into the Transient Structural module of ANSYS WORKBENCH to obtain the single-cycle load values on all finite element nodes of the single-cycle lower-life component.
3. The method for probabilistic failure risk analysis of life-limited components according to any one of claims 1 or 2, characterized in that, Step two specifically includes: selecting the maximum value s of the single-cycle load at the finite element node. base As a single-cycle characteristic load; the load randomness index includes dispersion, random load distribution pattern and number of cycles in the whole life cycle determined according to working history data or airworthiness standards; the load history of all cycles in the whole life cycle is constructed by random sampling.
4. The method for probabilistic failure risk analysis of life-limited components according to claim 3, characterized in that, The random load s is obtained based on the single-cycle characteristic load and its dispersion; the load distribution law within the entire cycle is obtained based on the single-cycle characteristic load and its random distribution law.
5. The method for probabilistic failure risk analysis of life-limited components according to claim 3, characterized in that, Step three specifically includes: selecting the transcendence curve of the initial crack distribution of the life-limited component; dividing the dimensional range containing the transcendence curve of the initial crack distribution into x intervals at equal intervals; labeling each x interval with an interval number d, where d = 1, 2, 3, ..., x; and recording the crack length at the left endpoint of each interval after equal interval division. and the length of the crack at the right end Calculate the crack length at the left end of each interval. and the length of the crack at the right end The probability of falling in interval d 6. The method for probabilistic failure risk analysis of life-limited components according to claim 5, characterized in that, Step four specifically includes: selecting the Wheeler load interference model with the built-in Paris formula, and obtaining the crack length at the right end of each interval based on the load history of the entire life cycle obtained in step two. and the length of the crack at the right end Correspondingly, the final value of crack length propagation at the left end of each interval after the full life cycle loading history. and the final value of the crack length propagation at the right end 7. The method for probabilistic failure risk analysis of life-limited components according to claim 6, characterized in that, Step five specifically includes: calculating the probability that the final value of the crack length extension at each interval's endpoint falls within interval d, based on the principle of probability conservation. middle.
8. The method for probabilistic failure risk analysis of life-limited components according to claim 7, characterized in that, Step six specifically includes: selecting the fracture toughness K. c As a criterion, the crack length a is obtained. critical ; Extend the final value of the crack length at the left and right endpoints of each interval. Determining the crack length a critical Compared to, obtaining satisfaction The interval number corresponding to the condition is used to generate the failure risk interval number j; if Then the failure risk P f =0, there is no failure risk interval index j; If an interval index j exists, then the total failure risk P f for: in, Let x be the probability value corresponding to the k-th interval, and let x be the maximum value of the interval index.
9. The method for probabilistic failure risk analysis of life-limited components according to any one of claims 1 or 2, characterized in that, The lifespan-limited component is the compressor wheel.
10. The method for probabilistic failure risk analysis of life-limited components according to claim 9, characterized in that, The compressor disc is made of titanium alloy.